Research On Dynamic Modeling And Control Of Carangiform Robotic Fish

This paper studies the dynamic modeling and target tracking control for a carangiform robotic fish. Based on the complete mathematical model of robotic fish, control oriented model processing methods are presented. In order to improve the disturbance attenuation and limit the amplitude of input, the prediction control is applied to the robotic fish control for the first time. Two target tracking control methods based on dynamic matrix control(DMC) and nonlinear model predictive control(NMPC) are designed, respectively. The main contribution of this paper can be summarized as follows:(1) The mathematical model of carangiform robotic fish is built. Firstly, by the coordinate transformation, the kinematic model is built and Lighthill’s large amplitude elongated-body theory is applied to evaluate the tail actuation induced hydrodynamic force, based on which, the dynamic model is derived. Then for the oscillation signal introduced by the periodic oscillating caudal fin when the robotic fish swims, by fixing the swinging amplitude and frequency, approximate averaging and two-order Taylor’s formula are employed respectively to obtain two equivalent dynamic models with only one input. At last, in order to the target tracking control design, an equivalent kinematic model only with distance and angle as control variables is obtained with transformation of original kinematic model.(2) Based on dynamic matrix predictive control, a target tracking control method for carangiform robotic fish is proposed. In order to improve the robustness of the system and prevent the control input from frequent saturation, model predictive control is used to design the target tracking controller for the robotic fish. Firstly, by discretizing the target tracking model and constructing the index function, the nonlinear DMC target tracking controller is presented. Then, the effect of the controller parameters on the control performance is analyzed, and the parameters optimization method is given. Finally, the effectiveness of the controller is verified by simulation.(3) A target tracking control method using NMPC for carangiform robotic fish is presented. In DMC, the amount of computation grows radically with the increase of the forecast time. To solve this problem, NMPC is introduced. By combining the characteristics of the robotic fish model with uncertain relative degree and introducing the terminal constraint and input constraint to rebuild the index function, NMPC based target tracking controller is designed. Simulation results show the effectiveness of the controller. Finally, the two predictive control methods presented in this paper are compared and analyzed, which provides their respective application conditions.